Although attributions of external controllability were found to be related to emotional exhaustion burnout in this study, the remaining dimensions of causal attributions were not found to be associated with any of the outcome measures. However this may be due to measurement problems, particularly given the low alphas produced for the stability sub-scale and the modest alphas produced for the personal controllability and locus of causality sub-scales. It is important for research to develop more reliable ways of measuring attributions other than external controllability in the third person. The internal consistency of the adapted CDSII sub-scales may be improved by adding 1 further item to each of the present 3, since the internal consistency of a measure improves with more items (Endler & Parker 1990). Furthermore, there may also be important dimensional properties of attributions which were not measured by the adapted CDSII including the dimensions of globality and the perceived intentionality of aggressive behaviour, which may have relevance to the psychological impact of aggression on staff and the way care staff respond on an affective, cognitive and behavioural basis. There is also a need for further research to examine how the particular beliefs of staff in relation to challenging behaviours are formed, and whether attitudes towards aggression in general are important predictors of staff responding. For instance family and cultural attitudes towards anger and aggresssion more generally may determine how tolerant staff are towards aggressive behaviour exhibited by the service-users they are caring for. It is also possible that personal attitudes about aggression and personal experience of aggression in a staff member’s personal life may also be an important predictor of whether or not an individual chooses to work in learning disabilities settings where aggressive service-user behaviour takes place.
The type of training staff are provided with for managing aggression may have an important influence on beliefs. Although this variable was not found to have any relationship with causal attributions in this study the measure of staff training for managing aggression needs much improvement since this measure did not take into account how long ago training occurred and did not involve any indication of the quality of training.
Although the present study looked at staff behefs in terms of attributions made about the cause of aggressive behaviour, there was no examination of beliefs about whether service user behaviour should be controlled, eliminated or tolerated. There is a need for research to examine how emotional responses and beliefs effect staff behaviour, motivation towards behavioural interventions and helping responses of staff and the caring relationship. This study did not examine what staff actually do in response to service-user aggression, particularly whether or not staff would intervene. Hastings (1995) found that 74% of care staff reported that they would intervene in response to aggressive service user behaviour. However, observational methods would be a more reliable way to explore this issue.
It is particularly important to further examine the relationship between emotional responses and staff behaviour given that Weiner (1980) found that emotional responses of anger were likely to inhibit helping responses. Anger is one of the emotions comprising the depression-related emotions scale used in this study and it would be important for future research to examine the role of depression- related versus anxiety-related emotional responses in determining staff behaviour and their respective roles in the maintenance of challenging behaviour.
The relationship between burnout and poor job performance has been assumed but not tested, and although the emotional responses of staff in the present study were found to predict emotional exhaustion and depersonalisation burnout, this only implies that negative emotional responses to service user aggression indirectly effects interactions between staff and service-users. Although Lawson and O ’Brien (1994) found a relationship between burnout and staff activity and behavioural
effects in learning disability settings, there is a need for more objective indices of job performance other than self-report inventories, such as observational methods.
Although path analysis can examine the pattern of relationships between variables, it cannot establish causal relationships between variables, there is a need for longitudinal research to establish causal links. Studies of stress and health outcome have largely relied on cross sectional designs (Chwalisz et al 1992), however, an exploration of the complex circular transactional model and complex relationships between variables require longitudinal research.
Situational versus Dispositional constructs:
This study investigated situational stress associated with exposure to service-user aggression and therefore relied on state measures of independent variables such as attributions, perceived self- efficacy and coping, rather that examine habitual or trait-like tendencies such as attributional 'style'
or coping 'style'. Carver et al (1989) are explicitly cautious about the relative validity of state or trait theories about these constructs and therefore developed the COPE inventory in a way that could be administered in situational or dispositional formats. It is also possible that certain members of staff are more depression-prone due to their more general beliefs. Hence dispositional measures of attributional style may be an important area for future research to examine in relation to depression- related emotional responses to service-user aggression.
Given that this study focused on situational responses to service user aggression the effects of more stable constructs such as personality and psychological type was not investigated in the present study. However there is some evidence that particular psychological types may be over-represented in the human services (Garden 1989), which has not yet been examined in learning disabilities settings. There is a possibility that different people are attracted to this kind of work due to stable personality characteristics which would suggest that the amount of aggression might therefore have no effects on outcome if this were the case.
This study also did not include a measure of recent life events experienced by staff which may produce high GHQ scores and therefore confound the results. There was also no measure of non- agression stressors related to the working environment, which may also have contributed to unexplained variance in outcomes. Although selection criteria specified that all participating staff were involved in direct care duties, the amount of hands on direct interaction between staff and service users may have varied from one staff member to another, according to shift patterns. Residential staff work varying shifts and staff who work night shifts may have less interaction with residents. There may also have been considerable diversity among participating staff in terms of grades and responsibilities.
There may also have been considerable diversity among the range of 23 participating staff teams on organisational factors that were not examined in this study. Variance due to diversity among teams was controlled for in this study in terms of the size of the staff team and the degree of learning disabilities and physical disabilities among residents cared for. But there may be other characteristics of the residents of the organisation which contribute to outcomes not measured in this study. For example, there may be powerful aspects of the service culture, such as rules about how to deal with service-user aggression, which may influence staff beliefs and outcomes (Hastings 1995). Staff rules may be generated by staff themselves or supplied more formally and may influence causal attributions. In order to investigate the contribution of organisational factors on the impact of aggression on staff, further research could compare reactions to aggressive service-user behaviour in diferent settings.
Implications of the study findings for staff training and support in services for people with learning disabilities who exhibit aggressive behaviour.
Given that the results of this study suggests that depression-related emotional responses to service- user aggression play a key role in mediating stress-related outcomes among staff caring for people with learning disabilities who exhibit aggression, there is a need for staff training and support to focus on helping staff manage these emotional responses more effectively. It appears that problem- focused and emotion-focused coping efforts as measured in this study are not sufficient to reduce the
frequency of negative emotional responses in relation to service-user aggression. However, the group of coping strategies referred to as ‘less adaptive coping’ may need to be discouraged among staff in favour of other more adaptive coping strategies which need clearly identifying in future research.
Staff training and support in the area of emotions is particularly important when considering the evidence the emotional reactions of staff exposed to aggressive behaviour are likely to influence the way a member of staff responds to the service-user. For example, fear/anxiety-related care staff emotional responses may result in avoidance responses and anger/depression-related emotions may result in punishing responses. As highlighted in the literature review at the beginning of this report, there is evidence that approach or avoidance behaviour in staff is influenced by the behaviour of service-users (Dailey et al 1974) [cited in Bromley and Emerson 1995] and there is evidence to suggest that these behavioural responses of staff may function to maintain rather than reduce challenging behaviours (Hastings & Remmington 1994). Given that negative emotional responses may therefore have important implications for a staff member’s ability to apply treatment programmes effectively it seems likely that staff training and support aimed at reducing negative emotional responses to service-user aggression may help to reduce aggressive service-user behaviour through the behaviour of staff. Furthermore, given the demonstrated contribution of negative emotional responses to more chronic staff outcomes, it is important that negative emotional responses to service user aggression are addressed and that staff are equipped to deal with these emotional responses.
The present finding that negative emotional responses to service-user aggression plays a key role in influencing emotional exhaustion burnout and depersonalisation burnout has important implications for helping staff with the emotional aspects of the work. However, in addition to anxiety and stress management training often provided for direct care staff, it is important that staff are helped to deal with depression-related emotional responses to service user aggression such as helplessness, sadness, anger and betrayal. Anger management training may be a particularly useful approach to take in addressing these emotional responses.
A form of stress inoculation training has recently been developed for staff who work with aggressive service-users (Keyes 1990) [cited in Hastings and Remmington 1994] which aims to teach staff skills to deal with this aversive aspect of their work. Information is provided to staff about the nature of stress and anger and the effects of their behaviour and staff are then taught coping skills, for example: self instruction on their behaviour, problem solving skills and relaxation techniques. Skills for managing aggresssion are then rehearsed in role playing exercises. This form of staff training was found to be effective in that less anger was reported afterwards in relation to service-user agression and less emergency restraint procedures were used. Given that anger is one of the depression-related emotional responses, this may be a very promising area of training for staff who are exposed to service-user aggression.
Furthermore, project managers involved in a PSSRU study in Kent by Knapp, Cambridge & Thomason (1989) identified coping with stress as an important training need in mental handicap services and only 52% of staff had received in-service training. However, in the light of the present study findings, comprehensive training packages aimed at helping staff cope with stress associated with service user aggression should address both coping strategies and negative emotional responses to aggressive service user behaviour.
The need for training for care staff dealing with aggressive or violent service-user behaviour has been emphasised in a DHSS report. Violence to Staff (1988) stating that ‘training has a key role
within the strategies against violence, and should be provided fo r all appropriate staff’ (cited in
McDonnell et al 1991a pp73) but it has not been clear what form training should take in learning disabilities services. Although many staff are trained in the use of control and restraint or breakaway techniques for breaking away from simple grips and holds, such practices are limited to managing situations after they have occurred. It is suggested that prevention of service user aggression and violence, by manipulating antecedents, is better than intervention. Unfortunately, many staff are not trained in the prevention of aggression related incidents. Staff need to be helped to develop skills for dealing with aggression and understanding it.
McDonnell et a/ (1991a) proposes that in services providing care for people with learning disabilities who exhibit aggression, the setting up of a training system is crucial, along with good reporting systems and clear unambiguous guidelines and policies for managing aggression effectively in order to minimise indecision and hesitation which may otherwise occur. McDonnell developed a model of training for care staff in learning disabilities services in health and social services settings, and highlighted the importance of taking into account the context of the setting, the nature and philosophy of the organisational system (Praill & Baldwin 1988; Wahler & Fox 1981).
Unfortunately, where an episode of service-user aggression or violence results in physical injury or distress, staff rarely have procedures for dealing with these effects. McDonnnell et a/ (1991a) states that 'trauma o f assault should be acknowledged by the caring system’ and proposes that counselling and support facilities should be readily available. Engel and Marsh (1986) have also emphasised the importance of support services for psychiatric staff who have been assaulted to help them cope and it seems equally important to provide opportunities for care staff in learning disabilities settings to work through emotional responses to episodes of service user aggression. The importance of opportunities for de-briefing after the event was highlighted by many participating staff in this study, in the space provided for comments at the end of the staff questionnaire and it seems likely that de-briefmg may play a crucial role in helping staff cope with the emotional impact of service- user aggression. However, as pointed out by one participant, it is important that de-briefing is constructive and not an inquest into what the staff member should have done.
The lack of support systems and training to deal with behaviour problems was found by George and Baumeister (1981) to be an important factor related to higher levels of staff turnover. Ward (1989) found that insufficient training and support for dealing with challenging behaviour was an important source of stress in community learning disabilities services. Social support is an important external factor which has been suggested to moderate the detrimental effects of violent service-user behaviour on staff (Wallis 1987) [cited in Whittington and Wykes 1992]. Support from other staff and managers has been found to be important in helping staff cope with distress associated with violent service-user incidents (see Rose 1993; Stenfert Kroese and Fleming 1992). However, demanding
relationships with colleagues has been found to be related to high staff stress (Rose 1993) and staff may need support from outside the organisation in order to feel more comfortable addressing sensitive emotional issues. Opportunities for staff to be supported through externally facilitated staff support groups might therefore be helpful, particularly in relation to helping staff deal with negative emotional responses to service-user aggression, since this study has demonstrated their importance in mediating stress-related outcomes among directs care staff caring for aggressive individuals. Support groups might be most appropriately facilitated by professionals in the learning disabilities field with both knowledge of challenging behaviour and skills in helping people cope with difficult emotions such as clinical psychologists.